1 I have tried to show this in "Maxims," Mind (1951).

2 Of some one appropriate logical type, if you are pedantic about counting such things as "characteristics". Professor Popper has pointed out to me that the argument does not depend on this finitude, anyway.

3 In actual fact the attitude of what I describe as genuine patriotism, and that of the abstracted, so to speak aesthetic enjoyment of the games-connoisseur frequently coexist, with the help of some double-think. I am talking of simplified types, and in saying what they can and cannot do am discussing what is and is not compatible with their definitions, and not of psychological possibility.

4 All these rules would of course be only formulable in a symbolism designed to accommodate imperatives as well as indicative statements. I am just assuming that such a symbolism is possible and that it would be fairly similar in its logic to the familiar symbolisms. I have not attempted to work it out.

5 For an interpretation of Hegel supporting these contentions see J. Hyppolite, Introduction a la Philosophie de l'Histoire de Hegel, M. Riviere et Cie., Paris, 1948.

6 This is notoriously a tricky piece of Kantian interpretation. I am for the sake of simplicity assuming the validity of the interpretation of Kant which makes him say that we are only free when doing our duty; an interpretation which is in accordance with one at any rate of Kant's uses of "free".

7 This means something like "varying one's behaviour in accordance with . . ." and not "pushes, in billiard ball fashion". It is sometimes argued that Kantian ethics presuppose a philosophy of mind construing motives as efficient causes, and that this invalidates his ethical theory. Neither the premiss nor the inference of this argument is valid.

8 Which of these represents what Kant intended to say is, again, a matter of controversial Kantian interpretation. The weaker and second interpretation gives rise to fewer difficulties and can be defended with reference to the text, but I nevertheless prefer the first as being more interesting, illuminating and characteristically Kantian, in that it follows obviously from typical Kantian premises and shows us where they lead. If Kant were merely another intuitionist, he would hardly deserve the attention he receives.

9 Not much should be made of this naturalness, for it will be relative to the language employed, i.e., to the kind of action-concepts occurring in it. When definitions of action-concepts are constructed, the definiens will be a concatenation of characteristics which, suitably phrased, can become parts of a maxim. But, as shown, the possibility of universalisation depends on which elements are included in the maxim. When we wish to include a particular set of these in a maxim, we can of course do this very simply if that set happens to correepond to the definiens of some natural action-word, by just including the name that action. Otherwise we must either invent a name and construct its definition, or include all the features which would have figured in that definition in the maxim itself.

10 Necessarily, for in the Kantian scheme there are only two alternatives for the groundings of a part or whole of the contents of a maxim; either to be entailed by the very form of the moral law, or to be empirically founded.

11 It is of course not made unambiguously clear in Kant's ethics whether he meant the formal principle to determine all the important details, or not. It really amounts to whether he meant to expound an ethic of obligation or one of permissibility whether he thought that morality was like Italy, where everything not forbidden is allowed, or like Germany, where everything not allowed was forbidden. I feel inclined to say that he tried at least to be on the German side, mainly because that makes for an interpretation which is simultaneously more difficult to defend philosophically and (a frequent conjunction, this, in issues of Kantian exegesis) more interesting, in that it more clearly follow from the general theses inspiring Kantian ethics, such as that of the "heteronomy" of all empirically motivated conduct.

12 ln a way, Kierkegaard was to metaphysical optimism what "Brave New World" is to scientific optimism.